Why is 60/40 supposed to be the best for retirement?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
vineviz
Posts: 13332
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why is 60/40 supposed to be the best for retirement?

Post by vineviz »

Logan Roy wrote: Wed Sep 21, 2022 7:58 pm What I said: 'if assets are providing $100 income a year, and liabilities are subtracting $90', then next year, I've got $10 more in assets, providing more income.
If you said that I'd say you are still confusing assets, liabilities, income, and cash flow. You're trying to draw a line connecting "assets" with "income" and "liabilities" with "expense", but there is only a partial connection.

I can have more assets than liabilities AND at the same time have more expenses than income. And vice versa.

A basic "Introduction to Financial Accounting" class would clear all this up for you much more completely than I can here.

The "path to wealth" is simply to have less consumption (or expenses) than you have income. Income minus consumption already has a name: "savings".
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
chassis
Posts: 1006
Joined: Tue Mar 24, 2020 4:28 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by chassis »

Hey @Logan Roy, what model and year car do you drive? When did you buy it?
Logan Roy
Posts: 420
Joined: Sun May 29, 2022 10:15 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by Logan Roy »

vineviz wrote: Wed Sep 21, 2022 8:10 pm
Logan Roy wrote: Wed Sep 21, 2022 7:58 pm What I said: 'if assets are providing $100 income a year, and liabilities are subtracting $90', then next year, I've got $10 more in assets, providing more income.
If you said that I'd say you are still confusing assets, liabilities, income, and cash flow. You're trying to draw a line connecting "assets" with "income" and "liabilities" with "expense", but there is only a partial connection.

I can have more assets than liabilities AND at the same time have more expenses than income. And vice versa.

A basic "Introduction to Financial Accounting" class would clear all this up for you much more completely than I can here.

The "path to wealth" is simply to have less consumption (or expenses) than you have income. Income minus consumption already has a name: "savings".
I've been a wealth manager and business analyst. And I don't have the problems you seem to with Kiyosaki's terminology. I think if you took a course in 'Basic Reading Comprehension', you probably wouldn't either. :beer
Logan Roy
Posts: 420
Joined: Sun May 29, 2022 10:15 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by Logan Roy »

chassis wrote: Wed Sep 21, 2022 8:25 pm Hey @Logan Roy, what model and year car do you drive? When did you buy it?
Drop me a Private Message if you want to talk cars.
chassis
Posts: 1006
Joined: Tue Mar 24, 2020 4:28 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by chassis »

Logan Roy wrote: Wed Sep 21, 2022 8:37 pm
chassis wrote: Wed Sep 21, 2022 8:25 pm Hey @Logan Roy, what model and year car do you drive? When did you buy it?
Drop me a Private Message if you want to talk cars.
Please post it here. You brought it up.
User avatar
vineviz
Posts: 13332
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why is 60/40 supposed to be the best for retirement?

Post by vineviz »

Logan Roy wrote: Wed Sep 21, 2022 8:33 pm I've been a wealth manager and business analyst. And I don't have the problems you seem to with Kiyosaki's terminology.
I have a problem with his terminology because it is wrong.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Charon
Posts: 547
Joined: Thu May 03, 2018 12:08 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by Charon »

Logan Roy wrote: Wed Sep 21, 2022 2:05 pm Well a quick recap: I'm representing Robert Kiyosaki's assertion that your house is better thought of as a liability, rather than an asset:
https://www.richdad.com/is-house-an-asset

I say he's right.
In fact you're both wrong, because the word "asset" is already defined, and it means something other than what you and Kiyosaki say. Quick primer: https://www.investopedia.com/terms/a/asset.asp . A helpful passage here is:

"An asset is, therefore, something that is owned by you or something that is owed to you. A $10 bill, a desktop computer, a chair, and a car are all assets. If you loaned money to someone, that loan is also an asset because you are owed that amount. For the person who owes it, the loan is a liability."

You can't just redefine words that are already have a standard meaning. As I said in my original post here, the definition you (and Kiyosaki) are proposing is nonsense. He says "what makes something an asset is that it puts money in your pocket each month". That means my retirement accounts are not assets. My CDs and I-bonds aren't assets. None of this stuff is putting money into my pocket each month - none of it is even accessible to me right now.

This is why people talk about liquid v. illiquid assets. Kiyosaki seems to be arguing that illiquid assets aren't assets. And even more than that, he argues that cash not earning interest isn't an asset (not putting money in your pocket...). Only a taxable investment account is an asset, and even then only when the market is going up...?

I think it's great to focus on living within one's means, but don't overgeneralize. My parents' house was the primary asset that funded their retirement. Their house was a net asset (value of the asset greater than the mortgage, the liability associated with the asset) the entire time they owned it.

But sometimes even negative net worth makes sense. Someone going into debt to go to medical school will end up far richer than I am (went into education, for my sins...).
seajay
Posts: 804
Joined: Sat May 01, 2021 3:26 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by seajay »

vineviz wrote: Tue Sep 20, 2022 3:06 pm
Poe22 wrote: Tue Sep 20, 2022 2:57 pm Just wanted to let you know, I've come across a page from Big ERN, where he actually mentions using cash/gold for the "safe part" of the transition portfolio in the "comment" sector, not just bonds :happy He describes gold results as promising!
Unfortunately, he's probably wrong on this (as he is on a couple different topics).

For one thing, gold is not a "safe" asset in any sense of that word. It has a highly volatile price and produces no predictable stream of income.
... OR cash (T-Bills) - a fiat currency, backed by nothing other than faith, are highly volatile in terms of gold (non-fiat commodity currency).

PV

A asset that can maintain purchase power over centuries without having to pay interest to (hopefully) offset the 2%/year target devaluation (inflation) rate, can be a distinct advantage. Especially if the asset may incur no ongoing costs once purchased - such as buried gold.

I'd agree with a equal blend of both fiat and non-fiat is better than either alone, simply on the grounds that concentration risk is a major risk factor and diversification is better than not.

Cross-link viewtopic.php?p=6883075#p6883075 that may be of interest. Much of outcome from whether a investor opts for stock/bonds or stock/gold is more down to luck. The more neutral stance is to hold a blend of both.
Topic Author
Poe22
Posts: 43
Joined: Thu Sep 15, 2022 1:21 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by Poe22 »

seajay wrote: Thu Sep 22, 2022 3:42 pm I'd agree with a equal blend of both fiat and non-fiat is better than either alone, simply on the grounds that concentration risk is a major risk factor and diversification is better than not.

Cross-link viewtopic.php?p=6883075#p6883075 that may be of interest. Much of outcome from whether a investor opts for stock/bonds or stock/gold is more down to luck. The more neutral stance is to hold a blend of both.
Backtests since 1971 agree with that: 60/20/20 (stocks/bonds/gold), or other variations including gold, outperforms 60/40. Cash doesn't help much in the portfolio.
https://www.portfoliovisualizer.com/bac ... tion4_3=10

Relying on backtesting, of course, always remains a risky thing.
ncbill
Posts: 1520
Joined: Sun Jul 06, 2008 4:03 pm
Location: Western NC

Re: Why is 60/40 supposed to be the best for retirement?

Post by ncbill »

Logan Roy wrote: Wed Sep 21, 2022 7:39 pm
vineviz wrote: Wed Sep 21, 2022 4:56 pm
Logan Roy wrote: Wed Sep 21, 2022 2:05 pm Well a quick recap: I'm representing Robert Kiyosaki's assertion that your house is better thought of as a liability, rather than an asset:
https://www.richdad.com/is-house-an-asset

I say he's right. But there is context. His central tenet is that the only reliable way to become wealthy is to maintain more in assets than liabilities. Studies on high net worth individuals show this pattern of financial behaviour forecasts wealth much better than salary or education. So if assets are providing $100 income a year, and liabilities are subtracting $90, then your wealth (and future income) is effectively compounding, year on year.
I've never read anything by Kiyosaki, but if the above is an accurate representation of anything he's written then I can't imagine why anyone would pay attention to him.

Maintaining "maintain more in assets than liabilities" is certainly not a path to wealth. And he (or you in the retelling) seems to be mixing up the concepts of cash flow, income, and net worth.
Logan Roy wrote: Wed Sep 21, 2022 2:05 pm But in my case, Kiyosaki's argument was completely self-evident, in my 20s, and buying a house then would've been a huge mistake.
Maybe. But renting an apartment that cost 110% of your gross income would ALSO have probably been a "huge mistake".

"Don't pay more for housing than you can afford" is generally decent financial advice.

"Your house is a liability" is just uninformed nonsense.
The problem is you've not given me anything of substance to respond to. You reject Kiyosaki. You reject the idea of assets vs liabilities being a path to wealth. I'm confused. He's confused. 'Nonsense'. But beyond a broad rejection, I'm none the wiser as to why he's wrong, or why you might be.

I worked full-time for 10 years. I invested 80% of my income, and averaged a 10% annual return. So by the age of 31, my portfolio was roughly 14x my annual salary (all in ISAs, tax-sheltered), and I never needed to work another day. I kept my liabilities very low. Some of the smartest people I know in machine learning are doing the same today – they're on high 6-figure salaries, and live on canals, in 'tiny homes', in a mobile home. This isn't on Kiyosaki's advice – this is actually fear these jobs won't exist in 10 years time.

I'm not hardline on renting vs owning. But rent and car leasing is covered today by 20% of my annual portfolio income. And I only needed to work for 10 years. If I'd been lucky enough to do this through the recent bull market, I'd have only needed to work 5 years. If any of us had been keen to get a mortgage, or buy a Mercedes, we'd be in the same situation 99% of people are (tied to work, and not accumulating much).
Kiyosaki is just another 'power of positive thinking' guru...cheaper & better reads in that genre are Dale Carnegie & Norman Vincent Peale.

That said, I agree that most people, even here, seem to want to fool themselves as to what a house truly costs in ongoing expenses.

When I see posts here in the vein of "I'm worth $4 million...a $3 million house & $1 million portfolio" I cringe a little.

A house is a highly illiquid asset with ridiculous transaction costs compared to equities & fixed income.

Grew up in what is now a ~$2 million house, watched mom blow through a hefty, tax-free property settlement (today's value ~$1 million) in just a few years after her divorce trying to keep up with a large (5BR/5BA) high-maintenance (WWI-era) house.

So today I live in a ~$200k townhouse half that size & pay a modest monthly fee that covers all outside maintenance, plus water/sewer.
seajay
Posts: 804
Joined: Sat May 01, 2021 3:26 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by seajay »

Poe22 wrote: Thu Sep 22, 2022 11:33 pm
seajay wrote: Thu Sep 22, 2022 3:42 pm I'd agree with a equal blend of both fiat and non-fiat is better than either alone, simply on the grounds that concentration risk is a major risk factor and diversification is better than not.

Cross-link viewtopic.php?p=6883075#p6883075 that may be of interest. Much of outcome from whether a investor opts for stock/bonds or stock/gold is more down to luck. The more neutral stance is to hold a blend of both.
Backtests since 1971 agree with that: 60/20/20 (stocks/bonds/gold), or other variations including gold, outperforms 60/40. Cash doesn't help much in the portfolio.
https://www.portfoliovisualizer.com/bac ... tion4_3=10

Relying on backtesting, of course, always remains a risky thing.
TSM/total world/PM versus TSM/world/T-Bills and they both had backtest/historic worst case %WR's of around similar values, but that occurred at different times. Such that a blend of TSM/world and 50/50 T-Bills/PM was better than either PM or T-Bills alone

Image

For that test, PM (precious metal) = silver from 1934, as a proxy/alternative to gold. In a pre 1933 world when money could be swapped for gold at a fixed/pegged rate it made more sense to hold T-Bills for their interest as that was like the state paying you for it to securely store your gold.

Similarly historically a blend of TSM/world was better than either alone, from a perspective of worst case 30 year % withdrawal rates.

Fundamentally all assets have their worst case outcomes, even all-world, diversification has a tendency for the worst case outcomes to not coincide in time, or for one to have a lower magnitude than the other even if they do coincide. Reducing the worst case is paramount to improving SWR.

Whilst investment gold was outlawed in America 1933 to 1970's, a investor could have opted to swap their gold into T-Bills, or Silver (poor mans gold). Elsewhere where investment gold was still permitted to be held broadly gold and silver tracked each other. At one time I did propose that PM (T-bills pre 1934, silver from 1934, and perhaps gold from mid 1970's) be added as yet another data option into Simba's backtest spreadsheet, however that seemingly was considered too contentious.
pitagora
Posts: 10
Joined: Thu Oct 29, 2020 3:14 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by pitagora »

dbr wrote: Fri Sep 16, 2022 9:01 am Right. 60/40 has been the location of a historical small optimum in safe withdrawal rate because SWR depends on avoiding worst worse cases and that works with 60/40 in 1929. When nothing worked was the late '60s and that tends to be the defining case for SWR. If a person's objectives are not defined by SWR it is hardly likely 60/40 will be uniquely better than other choices. It is reasonable to contemplate how one would like what a 100/0 portfolio might do, a 0/100 portfolio, or a 50/50 portfolio, or any of the options in between.
How do you know how these performed before 1979? Unfortunately portfolio backtester only goes that far, which like somebody else pointed out is a period with falling inflation and rates, ideal for bonds. I'd love to see how this works starting from 1929. Is there any backtesting tool (even if not free) that goes back further?
dbr
Posts: 42443
Joined: Sun Mar 04, 2007 9:50 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by dbr »

pitagora wrote: Tue Sep 27, 2022 2:59 am
dbr wrote: Fri Sep 16, 2022 9:01 am Right. 60/40 has been the location of a historical small optimum in safe withdrawal rate because SWR depends on avoiding worst worse cases and that works with 60/40 in 1929. When nothing worked was the late '60s and that tends to be the defining case for SWR. If a person's objectives are not defined by SWR it is hardly likely 60/40 will be uniquely better than other choices. It is reasonable to contemplate how one would like what a 100/0 portfolio might do, a 0/100 portfolio, or a 50/50 portfolio, or any of the options in between.
How do you know how these performed before 1979? Unfortunately portfolio backtester only goes that far, which like somebody else pointed out is a period with falling inflation and rates, ideal for bonds. I'd love to see how this works starting from 1929. Is there any backtesting tool (even if not free) that goes back further?
Tools like FireCalc or this one https://engaging-data.com/visualizing-4-rule/ go back to 1871. It is a fair question how well data that old is commensurate with current stocks and bonds. Here is a reference to Trinity: https://www.aaii.com/journal/199802/feature.pdf which goes back to 1926 and also has a table from 1946.
Topic Author
Poe22
Posts: 43
Joined: Thu Sep 15, 2022 1:21 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by Poe22 »

seajay wrote: Sun Sep 25, 2022 5:53 pm TSM/total world/PM versus TSM/world/T-Bills and they both had backtest/historic worst case %WR's of around similar values, but that occurred at different times. Such that a blend of TSM/world and 50/50 T-Bills/PM was better than either PM or T-Bills alone

Image

For that test, PM (precious metal) = silver from 1934, as a proxy/alternative to gold. In a pre 1933 world when money could be swapped for gold at a fixed/pegged rate it made more sense to hold T-Bills for their interest as that was like the state paying you for it to securely store your gold.

Similarly historically a blend of TSM/world was better than either alone, from a perspective of worst case 30 year % withdrawal rates.

Fundamentally all assets have their worst case outcomes, even all-world, diversification has a tendency for the worst case outcomes to not coincide in time, or for one to have a lower magnitude than the other even if they do coincide. Reducing the worst case is paramount to improving SWR.

Whilst investment gold was outlawed in America 1933 to 1970's, a investor could have opted to swap their gold into T-Bills, or Silver (poor mans gold). Elsewhere where investment gold was still permitted to be held broadly gold and silver tracked each other. At one time I did propose that PM (T-bills pre 1934, silver from 1934, and perhaps gold from mid 1970's) be added as yet another data option into Simba's backtest spreadsheet, however that seemingly was considered too contentious.
That's very interesting indeed. Which combination of assets (including non-stock/bond assets) have you found to be most promising?

I'm still trying to figure out if more diversification between asset classes (maximizing Sharpe) can also maximize SWR. Basically: Did some variation of a "risk parity", "permanent/all seasons" retirement portfolio outperform the classic 60/40 since 1871?

Of course, for accumulation it's 100% stocks for me, but I see that a retirement portfolio has different requirements due to regular withdrawals. I don't worry about volatility for retirement, but I do worry about SWR and worst portfolio durations (shortest time for the portfolio to go to zero). Of course, returns matter too, but only as second priority.
dbr
Posts: 42443
Joined: Sun Mar 04, 2007 9:50 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by dbr »

Poe22 wrote: Tue Sep 27, 2022 9:20 am
I'm still trying to figure out if more diversification between asset classes (maximizing Sharpe) can also maximize SWR. Basically: Did some variation of a "risk parity", "permanent/all seasons" retirement portfolio outperform the classic 60/40 since 1871?

You might be interested in the conversation at the end of this thread: viewtopic.php?t=386109&start=600
where I am asking exactly that sort of question.
Topic Author
Poe22
Posts: 43
Joined: Thu Sep 15, 2022 1:21 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by Poe22 »

dbr wrote: Tue Sep 27, 2022 11:14 am
Poe22 wrote: Tue Sep 27, 2022 9:20 am
I'm still trying to figure out if more diversification between asset classes (maximizing Sharpe) can also maximize SWR. Basically: Did some variation of a "risk parity", "permanent/all seasons" retirement portfolio outperform the classic 60/40 since 1871?

You might be interested in the conversation at the end of this thread: viewtopic.php?t=386109&start=600
where I am asking exactly that sort of question.
Thanks, that's actually quite helpful. It's so difficult sometimes filtering understandable and actionable information from conversations. I'd love to see more precise language regarding measures (CAGR, Sharpe, Correlation, Median, Average etc.). Sometimes I feel people are talking about different things, misunderstanding each other.

My most urgent questions are:

1. Should accumulation portfolios be constructed differently than withdrawal portfolios (for retirement)?
2. If so, what should a withdrawal portfolio be optimized for? Sharpe ratio? Asset correlation?
3. If so, does a higher Sharpe ratio and/or lower Correlation increase SWR?

I've already tried to address these questions here viewtopic.php?p=6883083#p6883083, but still know nothin after reading the responses.
Last edited by Poe22 on Tue Sep 27, 2022 11:56 am, edited 2 times in total.
dbr
Posts: 42443
Joined: Sun Mar 04, 2007 9:50 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by dbr »

Poe22 wrote: Tue Sep 27, 2022 11:41 am
dbr wrote: Tue Sep 27, 2022 11:14 am
Poe22 wrote: Tue Sep 27, 2022 9:20 am
I'm still trying to figure out if more diversification between asset classes (maximizing Sharpe) can also maximize SWR. Basically: Did some variation of a "risk parity", "permanent/all seasons" retirement portfolio outperform the classic 60/40 since 1871?

You might be interested in the conversation at the end of this thread: viewtopic.php?t=386109&start=600
where I am asking exactly that sort of question.
Thanks, that's actually quite helpful. It's so difficult sometimes filtering understandable and actionable information from conversations. I'd love to see more precise language regarding measures (CAGR, Sharpe, Correlation, Median, Average etc.). Sometimes I feel people are talking about different things, misunderstanding each other.

My most urgent questions are:

1. Should accumulation portfolios be constructed differently than withdrawal portfolios (for retirement)?
2. If so, what should a withdrawal portfolio be optimized for? Sharpe ratio? Asset correlation?
3. If so, does a higher Sharpe ratio and/or Correlation increase SWR?

I've already tried to address these questions here viewtopic.php?p=6883083#p6883083, but still know nothin after reading the responses.
I bet you'll end up where I am, namely reading some of the books suggested and then coming back to see where we are. I agree there is a lot of frustration from mumbo-jumbo rather than straightforward presentation of the concepts.

And I am pretty sure that I am neither investment nor mathematically naive and it still takes a lot of grounding to see what these ideas are all about.
dbr
Posts: 42443
Joined: Sun Mar 04, 2007 9:50 am

Re: Why is 60/40 supposed to be the best for retirement?

Post by dbr »

A good example of what I am talking about might be the first part of Ch2 in Lhabitant's book. It is just a basic laying out of how to formulate the behavior of a portfolio in statistical language. This is exactly equivalent to the first chapter in a book on classical mechanics in physics where one lays out how kinematics is formulated by positing the motion of particles using continuous functions of time to describe the positions of particles, the first derivative to define velocity, and the second derivative to define acceleration. Then one introduces Newton's law to relate accelerations to forces and masses and ends up with motion described by second order differential equations.

The statistical description of return and risk of a portfolio of investments is, of course, one of the first things I learned to appreciate when thinking about investing, but it is nice to have a book that puts it at the beginning of Ch2 in clear terms.
Topic Author
Poe22
Posts: 43
Joined: Thu Sep 15, 2022 1:21 pm

Re: Why is 60/40 supposed to be the best for retirement?

Post by Poe22 »

dbr wrote: Tue Sep 27, 2022 11:51 am
Poe22 wrote: Tue Sep 27, 2022 11:41 am
dbr wrote: Tue Sep 27, 2022 11:14 am
Poe22 wrote: Tue Sep 27, 2022 9:20 am
I'm still trying to figure out if more diversification between asset classes (maximizing Sharpe) can also maximize SWR. Basically: Did some variation of a "risk parity", "permanent/all seasons" retirement portfolio outperform the classic 60/40 since 1871?

You might be interested in the conversation at the end of this thread: viewtopic.php?t=386109&start=600
where I am asking exactly that sort of question.
Thanks, that's actually quite helpful. It's so difficult sometimes filtering understandable and actionable information from conversations. I'd love to see more precise language regarding measures (CAGR, Sharpe, Correlation, Median, Average etc.). Sometimes I feel people are talking about different things, misunderstanding each other.

My most urgent questions are:

1. Should accumulation portfolios be constructed differently than withdrawal portfolios (for retirement)?
2. If so, what should a withdrawal portfolio be optimized for? Sharpe ratio? Asset correlation?
3. If so, does a higher Sharpe ratio and/or Correlation increase SWR?

I've already tried to address these questions here viewtopic.php?p=6883083#p6883083, but still know nothin after reading the responses.
I bet you'll end up where I am, namely reading some of the books suggested and then coming back to see where we are. I agree there is a lot of frustration from mumbo-jumbo rather than straightforward presentation of the concepts.

And I am pretty sure that I am neither investment nor mathematically naive and it still takes a lot of grounding to see what these ideas are all about.

Thanks, it feels nice not being completely alone with these questions.
Post Reply